S-EMBER - Multiple Choice: leaderboard
Streaming egocentric memory retrieval (S-EMBER, Wang et al., Meta, 2026): 9,448 questions about 388 hours of first-person video from Ray-Ban Meta glasses, each asked at a trigger time and answerable only from what the wearer saw earlier, with open models given 128 uniformly sampled frames and proprietary models their platform limits. Five-way multiple choice with distractors written by an LLM from wrong answers of baseline models; accuracy in percent, higher is better.
Metric: Accuracy (%; five options with LLM-written distractors drawn from models' wrong answers) (S-EMBER: 9,448 questions on 3,141 egocentric Ray-Ban Meta videos, answered from the stream up to the question's trigger time; open models at 128 uniformly sampled frames, proprietary models at their platform limits). Source: arxiv.org. Saturation forecast: Around May 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 54.41 | #91 |
| 2 | Qwen 3 VL 32B Instruct | 42.4 | #582 |
| 3 | GPT-5.4 | 41.82 | #121 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=s-ember-multiple-choice · How It Works · Data refreshed daily, snapshot 2026-11-12.